1. Field of the Invention
This invention relates to data backup and more particularly relates to managing multiple copy versions of data from a source volume. Specifically, this invention relates to dynamically selecting and maintaining target volumes over multiple copy versions in a data copy environment.
2. Description of the Related Art
Data copy operations and tools are extremely important in today's computing environment because that is primarily how data is communicated among applications and users. One central objective of copying data is to create backup copies of data in case of failure or for restoration of a previous copy version corresponding to a specific state of the data copy environment at a particular instant in time.
Over time, data copy utilities have been developed that allow data copy operations to be performed in less time than before. Significantly, many point-in-time data copy technologies, such as FlashCopy and SnapShot, are capable of creating a virtual copy of trillions of data (Terabytes) in a matter of minutes, or even fractions of minutes. Given the huge amounts of data created and stored for all types of data applications, from personal computing to high-end data mining and analyzing, it is very important to be able to back up this data and to back it up in a way that is minimally interruptive to the data processing applications.
One way in which data backup intrusions are minimized is by using pre-pair processing in the data replication applications. Pre-pair processing employs pre-selection of a set of target volumes (also referred to as backup volumes) for a pre-defined set of source volumes. Pre-selection of the target volumes performs the pairing of sources and targets outside of the copy window so that the copy pairs are created before any data copying is performed. In this way, the copy pairing does not use critical processing time on the source computer to determine which targets will be used to back up the datasets on the source volumes. However, pre-selection of the target volumes introduces certain challenges to the data copy operations.
In certain scenarios, it may be difficult to maintain multiple sets of records to describe different states for a single source pool (a set of source volumes). This is apparent when source volumes are either added to or removed from the source pool. Replication records may describe previous or current copy versions of the source pool while pre-pairing records may indicate copy pairs that may be used in future copy versions. When changes to the source pool occur, the replication records and pre-pairing records may need to be updated individually and/or reconciled with each other in order to properly track the copies of the source volumes, the availability of the target volumes, and so forth.
Another potential challenge arises with specific reference to creating copy versions after a change has occurred in either a source pool or a target pool. Likewise, a further challenge may become present in maintaining the various data copy records for multiple sequential copy versions in the data copy environment. This challenge is amplified as the number of copy versions that are maintained increases.
What are needed are an apparatus, system, and method that are capable of addressing the challenges presented in current data backup and data copy environments. Beneficially, such an apparatus, system, and method would specifically overcome the known problems related to maintaining multiple sets of records, handling changes in the data copy environment that affect the copy pairs in the pre-pairing and replication records, and maintaining the records over a plurality of copy versions of a source pool in the data copy environment.